Macromolecules July-August 1972
Volume 5, Number 4 0
Copyright 1972 by the American Chemical Society
Swelling, Crazing, and Cracking of an Aromatic Copolyether-S'ulfone in Organic Media R. P. Kambour,* E. E. Romagosa, and C. L. Gruner Polymer Chemistry Branch, Chemical Laboratory, General Electric Corporate Research and Deuelopment, Schenectady, New York. Receiued March 30, 1972
ABSTRACT: The: critical flexural strain eo for crazing or cracking has been determined for compression-molded bars of an aromatic copolyether -sulfone in air and in 40 different small-moleculeorganic liquids. Measurements of degree of swelling of the polymer by 39 liquids have been also made and the resultant lowered glass temperature has been measured in 30 cases. Plasticization is responsible for the lowering of e , from its value in air to the minimum observed. As expected, maximum degree of swelling occurs iri agents having values of solubility parameter 6 close to that of the polymer. However, over a very broad range of 6 swelling can be sufficient to lower Tgand e, markedly from their dry-state values. The broad susceptibilityto limited swelling is a general characteristic of glassy polymers of high T, caused by the excess free energy of the glassy state.
I
n a previous study,' ciritical strains were reported for crazing of poly(2,6-dimethyl-l,4-phenylene oxide) in contact with a wide spectrum of organic agents. The critical strain was shown to be reduced in proportion t o the reduction of T, through plasticization. Since degree of plasticization depends primarily o n degree of swelling and the latter depends primarily o n the solubility parameters of polymer and swelling agent, a correlation was found between crazing resistance in a given agent and its solubility parameter 6. Specifically, critical strain was found to be reduced to a minimum at the same position o n the ii spectrum where swelling reached a maximum. Moreover, this value of 6 coincided with that for the polymer as calculated using Small's group contribution method. F o r a variety of reasons it is of interest t o explore how well these simple correlations hold for the solvent crazing of other glassy polymers. Is the reduction in critical strain exclusively a function of the ease of flow of the glass? H o w well does 6 act as a correlating parameter for the limited swelling of other glassy polymers in a wider variety of organic agents? Conversely, d o entropic factors, specifically those relating t o molecular size effects, play a negligible role in the swelling of other polymers by srnall organic molecules, as seems to be true for poly(pheny1ene oxide)? Finally, how does the solvent crazing susceptibility profile of a given polymer differ from the profile of poly(pheny1ene oxide)? Is the profile simply shifted o n the 6 spectrum t o the extent that the 6's for the two polymers are different') Or does the profile shape differ as well? With these questions as a background, we have conducted a n extensive study of the swelling and crazing characteristics of polysulfone-another glassy polymer of substantially higher solubility parameter hut of physical characteristics similar to those of poly(pheny1ene oxide): high T,, high
crazing resistance in air, and a yield stress of 10,000 psi, roughly.
Experimental Section (A) Materials. The resin, obtained from the Westlake Plastics Co., was an alternating copolyether. poly(oxy-l,4-phenylenesulfonyl- 1,4-phenyleneoxy-1,4-phenyleneisopropylidene-1,4 - phenylene), often referred to simply as polysulfone. The repeat unit in the chain is shown in I. The as-received resin is said to have a %I
0
L
1
0
I
of 25,000 on the basis of viscosity measurements. MJM,, should be 2.1-2.2.* Resin density is 1.24 g/cc and its T , is 186°C as determined here by scanning calorimetry. Before molding into sheets for crazing studies, the resin pellets were held 4 hours or more at 125" in a circulating air oven to drive off moisture. Sheets 0.04-0.05-in. thick were compression molded between aluminum foils at about 210". After 5 min under pressure at this temperature, the press was cooled over a 5-min interval to 180" and the sheets were then removed. They exhibited firstorder birefringent colors, as was true for the poly(ary1ene oxide) sheets that were used in the previous study of solvent crazing resistance. In neither case were attempts made to remove the birefringence by annealing. Bars for crazing were cut from the sheets and their edges milled smooth. They were washed in soapy water, rinsed, dried with paper towels, and then handled only with gloves or tongs prior to being strained. (E) Critical Strains. Values of critical strain ec were determined with Bergen elliptical strain jigs (Figure These afford con____ (2) See R. N. Johnson, A . G. Farnham, R. A. Clendinning, W. F. Hale, and C. N. Merriam, J . Pol.vm. Sci., Part A - I , 5, 2375 (1967), for molecular weight characteristics of these resins. (3) R. L. Bergen, Jr., SPEJ., 18 (6), 667 (1962).
(1) G . A . Bernier and R. P. I
